AI that explains itself

Most users walk away from AI tools unsure of what the system actually did or why it made specific decisions. That's a design problem we can solve.

We're already seeing products that show their reasoning before they act, explain decisions in plain language, and let users course-correct when the AI gets it wrong. In 2026, this will only grow. The explainable AI market is expected to reach $33.2 billion by 2032, as people won't trust systems they can't understand. The difference between AI products people adopt and those they abandon may come down to one simple question: do users know what the system just created in response to their prompts?

Agentic UX and human-agent ecosystems

With 88% of business leaders planning to increase AI budgets for agentic capabilities, AI agents are becoming a strategic priority. This coincides with growing user acceptance and knowledge of where, how, and when these agents are actually useful.

The result is consolidation. Instead of managing dozens of granular agents, master agents will orchestrate specialized agents automatically based on task type, context, and importance. Now, designers must create experiences for these human-agent ecosystems, managing agent lifecycles, orchestrating handoffs, and determining when humans step in.

Dynamic interfaces generated on demand

Large language models like Gemini 3 Pro can now build interactive, tailored interfaces in real-time for each prompt. Research shows these interfaces matched human expert-designed work 44% of the time, which is remarkable given they are generated in seconds.

Instead of handing off fixed screens, designers will need to create constraints, safety rails, and evaluation criteria that guide how these model-driven interfaces behave.

Voice interfaces find their place

Voice has moved past the hype phase: in the US alone, 157.1 million people are expected to use voice assistants by the end of 2026. Design leaders can expect a rise in context-aware, multimodal experiences. These interfaces fluidly combine voice, touch, and visuals based on users' actions. Good UX design now considers when people’s hands are busy or their environment makes typing impractical.

Micro-interactions become the language

Remember when those little animations and hover effects were extras designers added at the end of a project, like icing on the cake? A button changing color, a progress bar filling up, a small celebration when you complete a task... we treated them like decorative touches.

These micro-interactions have become the way interfaces communicate with users, acknowledging actions without forcing people to stop and read confirmation messages. An interface without them feels lifeless, like talking to someone who never looks up while you're speaking.

AR moves from demo to daily use

I've sat through dozens of impressive AR demos at conferences over the years, seen some cool stuff, and then gone back to my office thinking "when would I actually use this in real work?”

That gap between impressive demo and daily tool is starting to close. Retail companies are letting you see furniture in your living room before you buy it, and design teams are walking through 3D models in the real spaces where those designs will live.

AR starts making sense when it solves a real problem better than looking at flat images on a screen, and more companies are finding those problems worth solving.

Personalization without crossing the line

Users want personalized experiences until they realize what the system needs to know to deliver them. Predicting what you need before you ask sounds helpful, but it requires constant behavior tracking to work.

Take Netflix's recommendation algorithm. It works well because users opted into a service built around personalization. Compare that to a retail website that tracks your browsing across dozens of unrelated sites to show you ads for products you looked at once. Same technology, completely different user perception of whether it's helpful or invasive.

The best enterprise software gives users control over how much the system adapts to them and makes privacy settings visible. 2026 may be the year companies figure out how to adapt to user preferences without invasive tracking.

Accessibility gets built in, not added on

Accessibility can't be an afterthought or a compliance checkbox anymore. In one of my latest Forbes articles, I discussed how accessible design can drive ROI and become a genuine competitive advantage.

When you design for cognitive differences from the start, everyone benefits. For instance, motion-sensitivity toggles help people with ADHD, and those who just get dizzy from animations.

I expect accessibility to become one of the defining topics in UX design over the coming years, given its potential to improve how we build products.

Cross-platform UX that works

Users constantly switch between their phones, tablets, laptops, and smartwatches. They commute to work, spend hours at the office, take lunch breaks, head home, and expect their experience to follow them.

True cross-platform UX means your work follows you wherever you go. Start a task on mobile, continue on desktop, and finish on tablet without thinking about it. We'll see this becoming a much bigger part of business conversations in 2026.